diff --git a/website/content/docs/platform/k8s/injector/annotations.mdx b/website/content/docs/platform/k8s/injector/annotations.mdx index 219b01b4177c..3937f4b1d889 100644 --- a/website/content/docs/platform/k8s/injector/annotations.mdx +++ b/website/content/docs/platform/k8s/injector/annotations.mdx @@ -28,7 +28,7 @@ them, optional commands to run, etc. - `vault.hashicorp.com/agent-image` - name of the Vault docker image to use. This value overrides the default image configured in the injector and is usually - not needed. Defaults to `hashicorp/vault:1.18.1`. + not needed. Defaults to `hashicorp/vault:1.18.2`. - `vault.hashicorp.com/agent-init-first` - configures the pod to run the Vault Agent init container first if `true` (last if `false`). This is useful when other init diff --git a/website/content/docs/platform/k8s/injector/index.mdx b/website/content/docs/platform/k8s/injector/index.mdx index eafb61addf50..96a3126ba060 100644 --- a/website/content/docs/platform/k8s/injector/index.mdx +++ b/website/content/docs/platform/k8s/injector/index.mdx @@ -190,6 +190,27 @@ The configuration map must contain either one or both of the following files: An example of mounting a Vault Agent configmap [can be found here](/vault/docs/platform/k8s/injector/examples#configmap-example). +### Injector telemetry + +Vault Agent Injector collects the following Prometheus metrics (along with the +default set of golang metrics): + +- `vault_agent_injector_request_queue_length` - A gauge of webhook requests in + the injector's queue. + +- `vault_agent_injector_request_processing_duration_ms` - A histogram of webhook + request processing times in milliseconds. + +- `vault_agent_injector_injections_by_namespace_total` - The total count of + Agent container injections by Kubernetes `namespace` and `injection_type`, + where `injection_type` is `init_only`, `sidecar_only`, or `init_and_sidecar`. + +- `vault_agent_injector_failed_injections_by_namespace_total` - The total count + of failed Agent Sidecar injections by Kubernetes `namespace`. + +Enable metrics collection in the injector by setting [`injector.metrics.enabled: +true`](/vault/docs/platform/k8s/helm/configuration#metrics) in the Helm chart. + ## Tutorial Refer to the [Injecting Secrets into Kubernetes Pods via Vault Helm